@caupulican/pi-ai 0.85.3 → 0.85.8

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
Files changed (36) hide show
  1. package/dist/bedrock-provider.d.ts +1 -0
  2. package/dist/bedrock-provider.d.ts.map +1 -1
  3. package/dist/bedrock-provider.js +1 -0
  4. package/dist/bedrock-provider.js.map +1 -1
  5. package/dist/bedrock-scope.d.ts +35 -0
  6. package/dist/bedrock-scope.d.ts.map +1 -0
  7. package/dist/bedrock-scope.js +141 -0
  8. package/dist/bedrock-scope.js.map +1 -0
  9. package/dist/env-api-keys.d.ts.map +1 -1
  10. package/dist/env-api-keys.js +4 -1
  11. package/dist/env-api-keys.js.map +1 -1
  12. package/dist/image-models.generated.js +2 -2
  13. package/dist/image-models.generated.js.map +1 -1
  14. package/dist/models.generated.d.ts +219 -49
  15. package/dist/models.generated.d.ts.map +1 -1
  16. package/dist/models.generated.js +243 -115
  17. package/dist/models.generated.js.map +1 -1
  18. package/dist/stream.d.ts.map +1 -1
  19. package/dist/stream.js +193 -87
  20. package/dist/stream.js.map +1 -1
  21. package/dist/utils/tool-repair/registry.d.ts +21 -0
  22. package/dist/utils/tool-repair/registry.d.ts.map +1 -1
  23. package/dist/utils/tool-repair/registry.js +27 -0
  24. package/dist/utils/tool-repair/registry.js.map +1 -1
  25. package/dist/utils/tool-repair/text-protocol-history.d.ts.map +1 -1
  26. package/dist/utils/tool-repair/text-protocol-history.js +4 -1
  27. package/dist/utils/tool-repair/text-protocol-history.js.map +1 -1
  28. package/dist/utils/tool-repair/text-protocol-live-filter.d.ts +5 -0
  29. package/dist/utils/tool-repair/text-protocol-live-filter.d.ts.map +1 -1
  30. package/dist/utils/tool-repair/text-protocol-live-filter.js +11 -0
  31. package/dist/utils/tool-repair/text-protocol-live-filter.js.map +1 -1
  32. package/dist/utils/tool-repair/text-protocol.d.ts +5 -1
  33. package/dist/utils/tool-repair/text-protocol.d.ts.map +1 -1
  34. package/dist/utils/tool-repair/text-protocol.js +49 -27
  35. package/dist/utils/tool-repair/text-protocol.js.map +1 -1
  36. package/package.json +60 -2
@@ -4639,44 +4639,6 @@ export const MODELS = {
4639
4639
  contextWindow: 1000000,
4640
4640
  maxTokens: 128000,
4641
4641
  },
4642
- "gemini-2.5-pro": {
4643
- id: "gemini-2.5-pro",
4644
- name: "Gemini 2.5 Pro",
4645
- api: "openai-completions",
4646
- provider: "github-copilot",
4647
- baseUrl: "https://api.individual.githubcopilot.com",
4648
- headers: { "User-Agent": "GitHubCopilotChat/0.35.0", "Editor-Version": "vscode/1.107.0", "Editor-Plugin-Version": "copilot-chat/0.35.0", "Copilot-Integration-Id": "vscode-chat" },
4649
- compat: { "supportsStore": false, "supportsDeveloperRole": false, "supportsReasoningEffort": false },
4650
- reasoning: true,
4651
- input: ["text", "image"],
4652
- cost: {
4653
- input: 1.25,
4654
- output: 10,
4655
- cacheRead: 0.125,
4656
- cacheWrite: 0,
4657
- },
4658
- contextWindow: 128000,
4659
- maxTokens: 64000,
4660
- },
4661
- "gemini-3-flash-preview": {
4662
- id: "gemini-3-flash-preview",
4663
- name: "Gemini 3 Flash Preview",
4664
- api: "openai-completions",
4665
- provider: "github-copilot",
4666
- baseUrl: "https://api.individual.githubcopilot.com",
4667
- headers: { "User-Agent": "GitHubCopilotChat/0.35.0", "Editor-Version": "vscode/1.107.0", "Editor-Plugin-Version": "copilot-chat/0.35.0", "Copilot-Integration-Id": "vscode-chat" },
4668
- compat: { "supportsStore": false, "supportsDeveloperRole": false, "supportsReasoningEffort": false },
4669
- reasoning: true,
4670
- input: ["text", "image"],
4671
- cost: {
4672
- input: 0.5,
4673
- output: 3,
4674
- cacheRead: 0.05,
4675
- cacheWrite: 0,
4676
- },
4677
- contextWindow: 128000,
4678
- maxTokens: 64000,
4679
- },
4680
4642
  "gemini-3.1-pro-preview": {
4681
4643
  id: "gemini-3.1-pro-preview",
4682
4644
  name: "Gemini 3.1 Pro Preview",
@@ -4715,6 +4677,25 @@ export const MODELS = {
4715
4677
  contextWindow: 200000,
4716
4678
  maxTokens: 64000,
4717
4679
  },
4680
+ "gemini-3.6-flash": {
4681
+ id: "gemini-3.6-flash",
4682
+ name: "Gemini 3.6 Flash",
4683
+ api: "openai-completions",
4684
+ provider: "github-copilot",
4685
+ baseUrl: "https://api.individual.githubcopilot.com",
4686
+ headers: { "User-Agent": "GitHubCopilotChat/0.35.0", "Editor-Version": "vscode/1.107.0", "Editor-Plugin-Version": "copilot-chat/0.35.0", "Copilot-Integration-Id": "vscode-chat" },
4687
+ compat: { "supportsStore": false, "supportsDeveloperRole": false, "supportsReasoningEffort": false },
4688
+ reasoning: true,
4689
+ input: ["text", "image"],
4690
+ cost: {
4691
+ input: 1.5,
4692
+ output: 7.5,
4693
+ cacheRead: 0.15,
4694
+ cacheWrite: 0,
4695
+ },
4696
+ contextWindow: 1000000,
4697
+ maxTokens: 64000,
4698
+ },
4718
4699
  "gpt-4.1": {
4719
4700
  id: "gpt-4.1",
4720
4701
  name: "GPT-4.1",
@@ -4897,10 +4878,10 @@ export const MODELS = {
4897
4878
  thinkingLevelMap: { "off": null, "minimal": "low", "xhigh": "xhigh", "max": "max" },
4898
4879
  input: ["text", "image"],
4899
4880
  cost: {
4900
- input: 1,
4901
- output: 6,
4902
- cacheRead: 0.1,
4903
- cacheWrite: 1.25,
4881
+ input: 0.2,
4882
+ output: 1.2,
4883
+ cacheRead: 0.02,
4884
+ cacheWrite: 0,
4904
4885
  },
4905
4886
  contextWindow: 1050000,
4906
4887
  maxTokens: 128000,
@@ -4935,14 +4916,33 @@ export const MODELS = {
4935
4916
  thinkingLevelMap: { "off": null, "minimal": "low", "xhigh": "xhigh", "max": "max" },
4936
4917
  input: ["text", "image"],
4937
4918
  cost: {
4938
- input: 2.5,
4939
- output: 15,
4940
- cacheRead: 0.25,
4941
- cacheWrite: 3.125,
4919
+ input: 2,
4920
+ output: 12,
4921
+ cacheRead: 0.2,
4922
+ cacheWrite: 0,
4942
4923
  },
4943
4924
  contextWindow: 1050000,
4944
4925
  maxTokens: 128000,
4945
4926
  },
4927
+ "grok-4.5": {
4928
+ id: "grok-4.5",
4929
+ name: "Grok 4.5",
4930
+ api: "openai-completions",
4931
+ provider: "github-copilot",
4932
+ baseUrl: "https://api.individual.githubcopilot.com",
4933
+ headers: { "User-Agent": "GitHubCopilotChat/0.35.0", "Editor-Version": "vscode/1.107.0", "Editor-Plugin-Version": "copilot-chat/0.35.0", "Copilot-Integration-Id": "vscode-chat" },
4934
+ compat: { "supportsStore": false, "supportsDeveloperRole": false, "supportsReasoningEffort": false },
4935
+ reasoning: true,
4936
+ input: ["text", "image"],
4937
+ cost: {
4938
+ input: 2,
4939
+ output: 6,
4940
+ cacheRead: 0.5,
4941
+ cacheWrite: 0,
4942
+ },
4943
+ contextWindow: 500000,
4944
+ maxTokens: 128000,
4945
+ },
4946
4946
  "kimi-k2.7-code": {
4947
4947
  id: "kimi-k2.7-code",
4948
4948
  name: "Kimi K2.7 Code",
@@ -5666,23 +5666,6 @@ export const MODELS = {
5666
5666
  contextWindow: 131072,
5667
5667
  maxTokens: 32768,
5668
5668
  },
5669
- "meta-llama/llama-4-scout-17b-16e-instruct": {
5670
- id: "meta-llama/llama-4-scout-17b-16e-instruct",
5671
- name: "Llama 4 Scout 17B 16E",
5672
- api: "openai-completions",
5673
- provider: "groq",
5674
- baseUrl: "https://api.groq.com/openai/v1",
5675
- reasoning: false,
5676
- input: ["text", "image"],
5677
- cost: {
5678
- input: 0.11,
5679
- output: 0.34,
5680
- cacheRead: 0,
5681
- cacheWrite: 0,
5682
- },
5683
- contextWindow: 131072,
5684
- maxTokens: 8192,
5685
- },
5686
5669
  "openai/gpt-oss-120b": {
5687
5670
  id: "openai/gpt-oss-120b",
5688
5671
  name: "GPT OSS 120B",
@@ -5734,23 +5717,23 @@ export const MODELS = {
5734
5717
  contextWindow: 131072,
5735
5718
  maxTokens: 65536,
5736
5719
  },
5737
- "qwen/qwen3-32b": {
5738
- id: "qwen/qwen3-32b",
5739
- name: "Qwen3-32B",
5720
+ "qwen/qwen3.6-27b": {
5721
+ id: "qwen/qwen3.6-27b",
5722
+ name: "Qwen3.6 27B",
5740
5723
  api: "openai-completions",
5741
5724
  provider: "groq",
5742
5725
  baseUrl: "https://api.groq.com/openai/v1",
5743
5726
  reasoning: true,
5744
5727
  thinkingLevelMap: { "minimal": null, "low": null, "medium": null, "high": "default" },
5745
- input: ["text"],
5728
+ input: ["text", "image"],
5746
5729
  cost: {
5747
- input: 0.29,
5748
- output: 0.59,
5749
- cacheRead: 0,
5730
+ input: 0.6,
5731
+ output: 3,
5732
+ cacheRead: 0.3,
5750
5733
  cacheWrite: 0,
5751
5734
  },
5752
5735
  contextWindow: 131072,
5753
- maxTokens: 40960,
5736
+ maxTokens: 16384,
5754
5737
  },
5755
5738
  },
5756
5739
  "huggingface": {
@@ -6240,6 +6223,24 @@ export const MODELS = {
6240
6223
  contextWindow: 1048576,
6241
6224
  maxTokens: 384000,
6242
6225
  },
6226
+ "deepseek-ai/DeepSeek-V4-Flash-0731": {
6227
+ id: "deepseek-ai/DeepSeek-V4-Flash-0731",
6228
+ name: "DeepSeek V4 Flash 0731",
6229
+ api: "openai-completions",
6230
+ provider: "huggingface",
6231
+ baseUrl: "https://router.huggingface.co/v1",
6232
+ compat: { "supportsDeveloperRole": false },
6233
+ reasoning: true,
6234
+ input: ["text"],
6235
+ cost: {
6236
+ input: 0.14,
6237
+ output: 0.28,
6238
+ cacheRead: 0,
6239
+ cacheWrite: 0,
6240
+ },
6241
+ contextWindow: 1048576,
6242
+ maxTokens: 384000,
6243
+ },
6243
6244
  "deepseek-ai/DeepSeek-V4-Pro": {
6244
6245
  id: "deepseek-ai/DeepSeek-V4-Pro",
6245
6246
  name: "DeepSeek V4 Pro",
@@ -8929,7 +8930,7 @@ export const MODELS = {
8929
8930
  },
8930
8931
  "deepseek-v4-flash": {
8931
8932
  id: "deepseek-v4-flash",
8932
- name: "DeepSeek V4 Flash 0731",
8933
+ name: "DeepSeek V4 Flash (New)",
8933
8934
  api: "openai-completions",
8934
8935
  provider: "opencode",
8935
8936
  baseUrl: "https://opencode.ai/zen/v1",
@@ -10058,6 +10059,23 @@ export const MODELS = {
10058
10059
  contextWindow: 1000000,
10059
10060
  maxTokens: 65536,
10060
10061
  },
10062
+ "qwen3.8-max": {
10063
+ id: "qwen3.8-max",
10064
+ name: "Qwen3.8 Max",
10065
+ api: "anthropic-messages",
10066
+ provider: "opencode-go",
10067
+ baseUrl: "https://opencode.ai/zen/go",
10068
+ reasoning: true,
10069
+ input: ["text", "image"],
10070
+ cost: {
10071
+ input: 2,
10072
+ output: 6,
10073
+ cacheRead: 0.25,
10074
+ cacheWrite: 2.5,
10075
+ },
10076
+ contextWindow: 1000000,
10077
+ maxTokens: 131072,
10078
+ },
10061
10079
  },
10062
10080
  "openrouter": {
10063
10081
  "ai21/jamba-large-1.7": {
@@ -10837,7 +10855,7 @@ export const MODELS = {
10837
10855
  },
10838
10856
  "deepseek/deepseek-v4-flash": {
10839
10857
  id: "deepseek/deepseek-v4-flash",
10840
- name: "DeepSeek: DeepSeek V4 Flash",
10858
+ name: "DeepSeek: DeepSeek V4 Flash 0423",
10841
10859
  api: "openai-completions",
10842
10860
  provider: "openrouter",
10843
10861
  baseUrl: "https://openrouter.ai/api/v1",
@@ -10904,7 +10922,7 @@ export const MODELS = {
10904
10922
  input: 0.3,
10905
10923
  output: 2.5,
10906
10924
  cacheRead: 0.03,
10907
- cacheWrite: 0.08333333333333334,
10925
+ cacheWrite: 0.0833333333333333,
10908
10926
  },
10909
10927
  contextWindow: 1048576,
10910
10928
  maxTokens: 65535,
@@ -10921,7 +10939,7 @@ export const MODELS = {
10921
10939
  input: 0.09999999999999999,
10922
10940
  output: 0.39999999999999997,
10923
10941
  cacheRead: 0.01,
10924
- cacheWrite: 0.08333333333333334,
10942
+ cacheWrite: 0.0833333333333333,
10925
10943
  },
10926
10944
  contextWindow: 1048576,
10927
10945
  maxTokens: 65535,
@@ -10989,7 +11007,7 @@ export const MODELS = {
10989
11007
  input: 0.5,
10990
11008
  output: 3,
10991
11009
  cacheRead: 0.049999999999999996,
10992
- cacheWrite: 0.08333333333333334,
11010
+ cacheWrite: 0.0833333333333333,
10993
11011
  },
10994
11012
  contextWindow: 1048576,
10995
11013
  maxTokens: 65535,
@@ -11023,7 +11041,7 @@ export const MODELS = {
11023
11041
  input: 0.25,
11024
11042
  output: 1.5,
11025
11043
  cacheRead: 0.024999999999999998,
11026
- cacheWrite: 0.08333333333333334,
11044
+ cacheWrite: 0.0833333333333333,
11027
11045
  },
11028
11046
  contextWindow: 1048576,
11029
11047
  maxTokens: 65536,
@@ -11040,7 +11058,7 @@ export const MODELS = {
11040
11058
  input: 0.25,
11041
11059
  output: 1.5,
11042
11060
  cacheRead: 0.024999999999999998,
11043
- cacheWrite: 0.08333333333333334,
11061
+ cacheWrite: 0.0833333333333333,
11044
11062
  },
11045
11063
  contextWindow: 1048576,
11046
11064
  maxTokens: 65536,
@@ -11091,7 +11109,7 @@ export const MODELS = {
11091
11109
  input: 1.5,
11092
11110
  output: 9,
11093
11111
  cacheRead: 0.15,
11094
- cacheWrite: 0.08333333333333334,
11112
+ cacheWrite: 0.0833333333333333,
11095
11113
  },
11096
11114
  contextWindow: 1048576,
11097
11115
  maxTokens: 65536,
@@ -11108,7 +11126,7 @@ export const MODELS = {
11108
11126
  input: 0.3,
11109
11127
  output: 2.5,
11110
11128
  cacheRead: 0.03,
11111
- cacheWrite: 0.08333333333333334,
11129
+ cacheWrite: 0.0833333333333333,
11112
11130
  },
11113
11131
  contextWindow: 1048576,
11114
11132
  maxTokens: 65536,
@@ -11125,7 +11143,7 @@ export const MODELS = {
11125
11143
  input: 1.5,
11126
11144
  output: 7.5,
11127
11145
  cacheRead: 0.15,
11128
- cacheWrite: 0.08333333333333334,
11146
+ cacheWrite: 0.0833333333333333,
11129
11147
  },
11130
11148
  contextWindow: 1048576,
11131
11149
  maxTokens: 65536,
@@ -11446,13 +11464,13 @@ export const MODELS = {
11446
11464
  reasoning: false,
11447
11465
  input: ["text"],
11448
11466
  cost: {
11449
- input: 0.13,
11450
- output: 0.39999999999999997,
11467
+ input: 0.09999999999999999,
11468
+ output: 0.32,
11451
11469
  cacheRead: 0,
11452
11470
  cacheWrite: 0,
11453
11471
  },
11454
11472
  contextWindow: 131072,
11455
- maxTokens: 128000,
11473
+ maxTokens: 16384,
11456
11474
  },
11457
11475
  "meta-llama/llama-4-maverick": {
11458
11476
  id: "meta-llama/llama-4-maverick",
@@ -11582,9 +11600,9 @@ export const MODELS = {
11582
11600
  reasoning: true,
11583
11601
  input: ["text"],
11584
11602
  cost: {
11585
- input: 0.25,
11586
- output: 1,
11587
- cacheRead: 0.049999999999999996,
11603
+ input: 0.27,
11604
+ output: 1.08,
11605
+ cacheRead: 0.054,
11588
11606
  cacheWrite: 0,
11589
11607
  },
11590
11608
  contextWindow: 204800,
@@ -11837,8 +11855,8 @@ export const MODELS = {
11837
11855
  reasoning: false,
11838
11856
  input: ["text", "image"],
11839
11857
  cost: {
11840
- input: 0.075,
11841
- output: 0.19999999999999998,
11858
+ input: 0.09375,
11859
+ output: 0.25,
11842
11860
  cacheRead: 0,
11843
11861
  cacheWrite: 0,
11844
11862
  },
@@ -11957,9 +11975,9 @@ export const MODELS = {
11957
11975
  reasoning: true,
11958
11976
  input: ["text", "image"],
11959
11977
  cost: {
11960
- input: 0.6,
11961
- output: 3.41,
11962
- cacheRead: 0.19999999999999998,
11978
+ input: 0.589,
11979
+ output: 2.48,
11980
+ cacheRead: 0.0992,
11963
11981
  cacheWrite: 0,
11964
11982
  },
11965
11983
  contextWindow: 262144,
@@ -13540,12 +13558,12 @@ export const MODELS = {
13540
13558
  input: ["text"],
13541
13559
  cost: {
13542
13560
  input: 0.07,
13543
- output: 0.28,
13561
+ output: 0.27,
13544
13562
  cacheRead: 0,
13545
13563
  cacheWrite: 0,
13546
13564
  },
13547
13565
  contextWindow: 262144,
13548
- maxTokens: 262144,
13566
+ maxTokens: 32768,
13549
13567
  },
13550
13568
  "qwen/qwen3-coder-flash": {
13551
13569
  id: "qwen/qwen3-coder-flash",
@@ -13641,13 +13659,13 @@ export const MODELS = {
13641
13659
  reasoning: false,
13642
13660
  input: ["text"],
13643
13661
  cost: {
13644
- input: 0.09999999999999999,
13662
+ input: 0.09,
13645
13663
  output: 1.1,
13646
- cacheRead: 0.07,
13664
+ cacheRead: 0,
13647
13665
  cacheWrite: 0,
13648
13666
  },
13649
13667
  contextWindow: 262144,
13650
- maxTokens: 262144,
13668
+ maxTokens: 16384,
13651
13669
  },
13652
13670
  "qwen/qwen3-next-80b-a3b-thinking": {
13653
13671
  id: "qwen/qwen3-next-80b-a3b-thinking",
@@ -13664,7 +13682,7 @@ export const MODELS = {
13664
13682
  cacheWrite: 0,
13665
13683
  },
13666
13684
  contextWindow: 262144,
13667
- maxTokens: 32768,
13685
+ maxTokens: 4096,
13668
13686
  },
13669
13687
  "qwen/qwen3-vl-235b-a22b-instruct": {
13670
13688
  id: "qwen/qwen3-vl-235b-a22b-instruct",
@@ -13709,13 +13727,13 @@ export const MODELS = {
13709
13727
  reasoning: false,
13710
13728
  input: ["text", "image"],
13711
13729
  cost: {
13712
- input: 0.13,
13713
- output: 0.52,
13730
+ input: 0.15,
13731
+ output: 0.6,
13714
13732
  cacheRead: 0,
13715
13733
  cacheWrite: 0,
13716
13734
  },
13717
13735
  contextWindow: 262144,
13718
- maxTokens: 32768,
13736
+ maxTokens: 16384,
13719
13737
  },
13720
13738
  "qwen/qwen3-vl-30b-a3b-thinking": {
13721
13739
  id: "qwen/qwen3-vl-30b-a3b-thinking",
@@ -13794,8 +13812,8 @@ export const MODELS = {
13794
13812
  reasoning: true,
13795
13813
  input: ["text", "image"],
13796
13814
  cost: {
13797
- input: 0.39999999999999997,
13798
- output: 3.1999999999999997,
13815
+ input: 0.26,
13816
+ output: 2.08,
13799
13817
  cacheRead: 0,
13800
13818
  cacheWrite: 0,
13801
13819
  },
@@ -13930,13 +13948,13 @@ export const MODELS = {
13930
13948
  reasoning: true,
13931
13949
  input: ["text", "image"],
13932
13950
  cost: {
13933
- input: 0.3,
13934
- output: 2,
13935
- cacheRead: 0.15,
13951
+ input: 0.28900000000000003,
13952
+ output: 2.4,
13953
+ cacheRead: 0,
13936
13954
  cacheWrite: 0,
13937
13955
  },
13938
13956
  contextWindow: 262144,
13939
- maxTokens: 65536,
13957
+ maxTokens: 131072,
13940
13958
  },
13941
13959
  "qwen/qwen3.6-35b-a3b": {
13942
13960
  id: "qwen/qwen3.6-35b-a3b",
@@ -14057,6 +14075,23 @@ export const MODELS = {
14057
14075
  contextWindow: 1000000,
14058
14076
  maxTokens: 131072,
14059
14077
  },
14078
+ "qwen/qwen3.8-max": {
14079
+ id: "qwen/qwen3.8-max",
14080
+ name: "Qwen: Qwen3.8 Max",
14081
+ api: "openai-completions",
14082
+ provider: "openrouter",
14083
+ baseUrl: "https://openrouter.ai/api/v1",
14084
+ reasoning: true,
14085
+ input: ["text", "image"],
14086
+ cost: {
14087
+ input: 2,
14088
+ output: 6,
14089
+ cacheRead: 0.25,
14090
+ cacheWrite: 2.5,
14091
+ },
14092
+ contextWindow: 1000000,
14093
+ maxTokens: 131072,
14094
+ },
14060
14095
  "rekaai/reka-edge": {
14061
14096
  id: "rekaai/reka-edge",
14062
14097
  name: "Reka Edge",
@@ -14263,7 +14298,7 @@ export const MODELS = {
14263
14298
  },
14264
14299
  "x-ai/grok-4.20": {
14265
14300
  id: "x-ai/grok-4.20",
14266
- name: "xAI: Grok 4.20",
14301
+ name: "SpaceXAI: Grok 4.20",
14267
14302
  api: "openai-completions",
14268
14303
  provider: "openrouter",
14269
14304
  baseUrl: "https://openrouter.ai/api/v1",
@@ -14280,7 +14315,7 @@ export const MODELS = {
14280
14315
  },
14281
14316
  "x-ai/grok-4.3": {
14282
14317
  id: "x-ai/grok-4.3",
14283
- name: "xAI: Grok 4.3",
14318
+ name: "SpaceXAI: Grok 4.3",
14284
14319
  api: "openai-completions",
14285
14320
  provider: "openrouter",
14286
14321
  baseUrl: "https://openrouter.ai/api/v1",
@@ -14297,7 +14332,7 @@ export const MODELS = {
14297
14332
  },
14298
14333
  "x-ai/grok-4.5": {
14299
14334
  id: "x-ai/grok-4.5",
14300
- name: "xAI: Grok 4.5",
14335
+ name: "SpaceXAI: Grok 4.5",
14301
14336
  api: "openai-completions",
14302
14337
  provider: "openrouter",
14303
14338
  baseUrl: "https://openrouter.ai/api/v1",
@@ -14314,7 +14349,7 @@ export const MODELS = {
14314
14349
  },
14315
14350
  "x-ai/grok-build-0.1": {
14316
14351
  id: "x-ai/grok-build-0.1",
14317
- name: "xAI: Grok Build 0.1",
14352
+ name: "SpaceXAI: Grok Build 0.1",
14318
14353
  api: "openai-completions",
14319
14354
  provider: "openrouter",
14320
14355
  baseUrl: "https://openrouter.ai/api/v1",
@@ -14542,9 +14577,9 @@ export const MODELS = {
14542
14577
  reasoning: true,
14543
14578
  input: ["text"],
14544
14579
  cost: {
14545
- input: 1.19,
14546
- output: 3.74,
14547
- cacheRead: 0.221,
14580
+ input: 0.76,
14581
+ output: 2.42,
14582
+ cacheRead: 0.14,
14548
14583
  cacheWrite: 0,
14549
14584
  },
14550
14585
  contextWindow: 1048576,
@@ -14670,7 +14705,7 @@ export const MODELS = {
14670
14705
  input: 1.5,
14671
14706
  output: 7.5,
14672
14707
  cacheRead: 0.15,
14673
- cacheWrite: 0.08333333333333334,
14708
+ cacheWrite: 0.0833333333333333,
14674
14709
  },
14675
14710
  contextWindow: 1048576,
14676
14711
  maxTokens: 65536,
@@ -14817,6 +14852,25 @@ export const MODELS = {
14817
14852
  contextWindow: 1000000,
14818
14853
  maxTokens: 384000,
14819
14854
  },
14855
+ "deepseek-v4-flash-0731": {
14856
+ id: "deepseek-v4-flash-0731",
14857
+ name: "DeepSeek V4 Flash 0731",
14858
+ api: "openai-completions",
14859
+ provider: "qwen-token-plan",
14860
+ baseUrl: "https://token-plan.ap-southeast-1.maas.aliyuncs.com/compatible-mode/v1",
14861
+ compat: { "thinkingFormat": "qwen", "supportsDeveloperRole": false, "supportsStore": false },
14862
+ reasoning: true,
14863
+ thinkingLevelMap: { "minimal": null, "low": null, "medium": null, "high": "high", "xhigh": "max" },
14864
+ input: ["text"],
14865
+ cost: {
14866
+ input: 0,
14867
+ output: 0,
14868
+ cacheRead: 0,
14869
+ cacheWrite: 0,
14870
+ },
14871
+ contextWindow: 1000000,
14872
+ maxTokens: 384000,
14873
+ },
14820
14874
  "deepseek-v4-pro": {
14821
14875
  id: "deepseek-v4-pro",
14822
14876
  name: "DeepSeek V4 Pro",
@@ -15016,6 +15070,24 @@ export const MODELS = {
15016
15070
  contextWindow: 1000000,
15017
15071
  maxTokens: 65536,
15018
15072
  },
15073
+ "qwen3.8-max": {
15074
+ id: "qwen3.8-max",
15075
+ name: "Qwen3.8 Max",
15076
+ api: "openai-completions",
15077
+ provider: "qwen-token-plan",
15078
+ baseUrl: "https://token-plan.ap-southeast-1.maas.aliyuncs.com/compatible-mode/v1",
15079
+ compat: { "thinkingFormat": "qwen", "supportsDeveloperRole": false, "supportsStore": false },
15080
+ reasoning: true,
15081
+ input: ["text", "image"],
15082
+ cost: {
15083
+ input: 0,
15084
+ output: 0,
15085
+ cacheRead: 0,
15086
+ cacheWrite: 0,
15087
+ },
15088
+ contextWindow: 1000000,
15089
+ maxTokens: 131072,
15090
+ },
15019
15091
  "qwen3.8-max-preview": {
15020
15092
  id: "qwen3.8-max-preview",
15021
15093
  name: "Qwen3.8 Max Preview",
@@ -15091,6 +15163,25 @@ export const MODELS = {
15091
15163
  contextWindow: 1000000,
15092
15164
  maxTokens: 384000,
15093
15165
  },
15166
+ "deepseek-v4-flash-0731": {
15167
+ id: "deepseek-v4-flash-0731",
15168
+ name: "DeepSeek V4 Flash 0731",
15169
+ api: "openai-completions",
15170
+ provider: "qwen-token-plan-cn",
15171
+ baseUrl: "https://token-plan.cn-beijing.maas.aliyuncs.com/compatible-mode/v1",
15172
+ compat: { "thinkingFormat": "qwen", "supportsDeveloperRole": false, "supportsStore": false },
15173
+ reasoning: true,
15174
+ thinkingLevelMap: { "minimal": null, "low": null, "medium": null, "high": "high", "xhigh": "max" },
15175
+ input: ["text"],
15176
+ cost: {
15177
+ input: 0,
15178
+ output: 0,
15179
+ cacheRead: 0,
15180
+ cacheWrite: 0,
15181
+ },
15182
+ contextWindow: 1000000,
15183
+ maxTokens: 384000,
15184
+ },
15094
15185
  "deepseek-v4-pro": {
15095
15186
  id: "deepseek-v4-pro",
15096
15187
  name: "DeepSeek V4 Pro",
@@ -15290,6 +15381,24 @@ export const MODELS = {
15290
15381
  contextWindow: 1000000,
15291
15382
  maxTokens: 65536,
15292
15383
  },
15384
+ "qwen3.8-max": {
15385
+ id: "qwen3.8-max",
15386
+ name: "Qwen3.8 Max",
15387
+ api: "openai-completions",
15388
+ provider: "qwen-token-plan-cn",
15389
+ baseUrl: "https://token-plan.cn-beijing.maas.aliyuncs.com/compatible-mode/v1",
15390
+ compat: { "thinkingFormat": "qwen", "supportsDeveloperRole": false, "supportsStore": false },
15391
+ reasoning: true,
15392
+ input: ["text", "image"],
15393
+ cost: {
15394
+ input: 0,
15395
+ output: 0,
15396
+ cacheRead: 0,
15397
+ cacheWrite: 0,
15398
+ },
15399
+ contextWindow: 1000000,
15400
+ maxTokens: 131072,
15401
+ },
15293
15402
  "qwen3.8-max-preview": {
15294
15403
  id: "qwen3.8-max-preview",
15295
15404
  name: "Qwen3.8 Max Preview",
@@ -15422,6 +15531,25 @@ export const MODELS = {
15422
15531
  contextWindow: 1000000,
15423
15532
  maxTokens: 500000,
15424
15533
  },
15534
+ "deepseek-ai/DeepSeek-V4-Flash-0731": {
15535
+ id: "deepseek-ai/DeepSeek-V4-Flash-0731",
15536
+ name: "DeepSeek V4 Flash 0731",
15537
+ api: "openai-completions",
15538
+ provider: "together",
15539
+ baseUrl: "https://api.together.ai/v1",
15540
+ compat: { "supportsStore": false, "supportsDeveloperRole": false, "supportsReasoningEffort": false, "maxTokensField": "max_tokens", "supportsStrictMode": false, "supportsLongCacheRetention": false, "thinkingFormat": "together" },
15541
+ reasoning: true,
15542
+ thinkingLevelMap: { "minimal": null, "low": null, "medium": null },
15543
+ input: ["text"],
15544
+ cost: {
15545
+ input: 0.14,
15546
+ output: 0.28,
15547
+ cacheRead: 0.03,
15548
+ cacheWrite: 0,
15549
+ },
15550
+ contextWindow: 1000000,
15551
+ maxTokens: 384000,
15552
+ },
15425
15553
  "deepseek-ai/DeepSeek-V4-Pro": {
15426
15554
  id: "deepseek-ai/DeepSeek-V4-Pro",
15427
15555
  name: "DeepSeek V4 Pro",